Last year, I wrote a blog series on Leuven Database of Ancient Books (LDAB), in which I queried the metadata of ancient manuscripts, including their bookform, material, manuscript counts, popularity of classical authors, and the historical formation of the New Testament Canon. To my pleasant surprise, that series became by far the most popular series in 2018.
In this post, I’ll look into another topic related to the New Testament, namely, manuscript coverage. Manuscript coverage is a measure of the reliability of a transcript.[1] There are many ways to measure manuscript coverage, but the basic idea is that the more manuscripts there are, and the earlier the manuscripts, that support a particular transcript, the more reliable is the transcript.
To quantify manuscript coverage, we need to divide the text into units. Although there were no chapter and verse divisions in the earliest manuscripts of the New Testament, chapters are helpful conceptual units to work with. To be more rigorous, I would need to work with verses, even individual words, but the coverage metadata for them are not readily available online.[2] The type of analysis I envision and describe here is applicable not only to the New Testament, but also to all ancient texts, provided that the relevant metadata are in a format readable by software programs.
Figure 1. New Testament Manuscript Coverage From 2nd to 3rd Century
There are a total of 260 chapters in the 27 books of the New Testament. 184 chapters (70% of the total) are covered by at least one manuscript by the end of 3rd century (Figure 1). Manuscript coverage is not evenly distributed among the New Testament books. The Gospel of John has the best coverage, as each of the 21 chapters of the book is covered by at least two manuscripts (John 1 is covered by five manuscripts). Many of Paul’s epistles have 1X coverage, with each chapter of the book covered by at least one manuscript. Matthew, Mark, and Revelation are half covered. 2 Timothy, 1 and 2 Peter, 2 and 3 John not at all.
A few early 3rd century manuscripts are notable in terms of coverage: P66 contains all chapters of John; P46 contains all of 1 and 2 Corinthians, Galatians, Philippians, Colossians, Ephesians, and Hebrews, and the majority of 1 Thessalonians (3 of 5 chapters) and Romans (11 of 16 chapters). P45 contains all four gospels and Acts, but no more than half of the chapters in each book.
Manuscript coverage of the New Testament was complete by the 4th century (Figure 2). Codex Sinaiticus (4th century) contains all 27 books (all 260 chapters) of the New Testament; Codex Vaticanus (4th) misses 1 and 2 Timothy, Titus and Philemon; Codex Alexandrinus (5th) contains all 27 books, but misses large parts of Matthew and 2 Corinthians; Codex Ephraemi Rescriptus (5th) misses 2 Thessalonians, 2 John and parts of other books. By the 8th century, average coverage is 10X (i.e., 10 manuscript per chapter). The Gospels as a whole have better coverage than the other books, perhaps suggesting their popularity. By the 19th century (not shown in the figure), average coverage of the New Testament is 300X.
Figure 2. New Testament Manuscript Coverage From 3rd to 8th Century
Notes:
^1. In the context of this series, I use the word “transcript” to refer to the content of the text, and “manuscript” the physical object containing the text.
^2. All the metadata I used for this blogpost are retrieved from New Testament Virtual Manuscript Room through the API provided by INTF (The Institut für neutestamentliche Textforschung), although the same metadata are also available from LDAB. I chose the former because it is ready-made, and, when the two disagree, it seems to be the more accurate.
References:
- New Testament Virtual Manuscript Room – INTF. Accessed March 7, 2019. http://ntvmr.uni-muenster.de.
- The Institut Für Neutestamentliche Textforschung. “The Bible Tool.” CrossWire Bible Society. Accessed March 07, 2019. http://crosswire.org/study/rb.